Fossils found along the Kotuikan River in northern Siberia doc-ument the Cambrian “Explosion,” the remarkable flowering ofanimal life that began some 543 million years ago. As CharlesDarwin recognized more than a century ago, Cambrian fossilsraise fundamental questions about life’s earlier evolution. Whatkind of organisms preceded these already complex animals? Canwe find older rocks, and if we can, will they preserve a record ofEarth’s earliest biological history?

Sometimes the past was shot with a hand-held camera; some-

times it reared monumentally inside a proscenium arch with

moulded plaster swags and floppy curtains; sometimes it

eased along, a love story from the silent era, pleasing, out of

focus and wholly implausible. And sometimes there was only

a succession of stills to be borrowed from the memory.

—Julian Barnes

Staring at the Sun

The cliffs along the Kotuikan River glow fawn and pink in the lateafternoon sun (figure 1.1). Elsewhere, in North America or in Europe, avista like this would be celebrated as a national park, its approachesflanked by campgrounds and souvenir shops. But here, in the forestedwilderness of northern Siberia, its pastel beauty is both unremarkableand largely unseen. From a sheltered niche halfway up the cliff, I lookup at my friend Misha Semikhatov perched high above the river, hislarge frame barely supported by a narrow ledge. The drop beneath his

1IntheBeginning?

feet is precipitous, but Misha’s attention is elsewhere, fixed on a layer ofsedimentary rocks just above his head. To his experienced eye, the bedof crinkly laminated limestones tells of an ancient tidal flat that bor-dered a vanished ocean, a broad expanse of shoreline exposed at lowtide, covered by thickly matted bacteria, and occasionally crossed bysmall animals. As I rest against the rock face, observing marginally

7In the Beginning?

Figure 1.1. Fossiliferous cliffs along the Kotuikan River in Siberia. The distancefrom river level to the top of the cliff is more than 300 feet, recording some 20million years of Early Cambrian history.

older beds, jotting in my notebook, and swatting mosquitoes (not nec-essarily in that order), I reflect on what has brought Misha and me tothis remote spot high above the Arctic Circle (figure 1.2). The literal an-swer is a giant Soviet-era military helicopter that deposited us, a smallgroup of colleagues, and a ton of gear some seventy miles upstream.From there, small rubber rafts floated us like Huckleberry Finn slowlydown the river, through canyons of limestone, beneath circling falcons,past wolves that howl at the midnight sun, to this wild and beautifulplace.

Of course, helicopters provide only one of several appropriate re-sponses to the question of what brought us here. The deeper and moreinteresting answer is that these cliffs, cut over the millennia by the Ko-tuikan as it winds toward the Arctic Ocean, record one of Earth history’sgreat turning points. As well as any rocks known anywhere, they docu-ment the remarkable diversification of animal life popularly known asthe Cambrian Explosion. In the broadest possible sense, the Kotuikancliffs record the beginnings of the modern world, a world in which an-imals swim, crawl, or walk beneath an atmosphere of breathable air.That’s really what brought us here.

At river level, a series of steps rise out of the water like prehistoricghats, hewn by nature from thin beds of limestone and dolomite. Some

8 Chapter 1

Burgess Shale XI

Akademikerbreen IIIKotuikan I, XI

Doushantuo IX

Warrawoona IV

Great Wall VII

Gunflint VI

Nama X

Figure 1.2. Map showing the location of the Kotuikan cliffs, along with princi-pal localities discussed in subsequent chapters (denoted by roman numerals).

545 million years ago, these rocks were deposited as lime muds in awarm shallow seaway not unlike the modern Florida Keys. Scatteredclusters of gypsum crystals record drying that episodically left coastalwaters salty enough to exclude all but the hardiest bacteria. The fossilsof animals are rare in these rocks, and those that can be found aresimple. Only a few irregular meanders disturb bedding surfaces, thetrails of small wormlike creatures that crawled along the muddy bottomin search of food.

About ten feet above the river, an abrupt shift to quartz sandstonemarks the so-called Precambrian-Cambrian1 boundary, historically theline of demarcation between the tractable paleontology of the Phanero-zoic Eon (literally, the “age of visible life”) and the terra incognita of amore youthful Earth. Volcanic rocks a few hundred miles to the eastdate this horizon at 543 (plus or minus one) million years before thepresent. Above the sandstone bench purple, red, and green shales forma steep shoulder above which vertiginous cliffs of limestone rise like awall. The shales record a flooding event, with shoreline sands pushedfar to the west by the rising sea. As sediments accumulated, the seaagain grew shallower, so that the overlying limestone beds record envi-ronments progressively closer to the ancient shore. Near the top of thecliff face, an irregular surface marks a point at which the sediments wereexposed and eroded by some vanished forebear of the Kotuikan River,only to be drowned again as the sea reclaimed lost territory.

Beginning at the level of the sandstone bench, the rocks containsmall skeletal fossils. In the lowermost beds, there are only a fewforms, hollow cones of calcite little more than a millimeter long (figure1.3a). But as we ascend the cliff, slowly and carefully to avoid a career-shortening slip, the abundance and variety of these fossils increase. So,too, do the number and behavioral complexity of preserved tracks,trails, and burrows. Near the top of the cliff, more than three hundred

9In the Beginning?

1 By convention, geologic time is divided into four eons: the Phanerozoic(0–543 million years ago), the Proterozoic (543–million 2.5 billion years ago), theArchean (2.5–ca. 4 billion years ago), and the Hadean (the time interval from theaccretion of the Earth to the beginning of the preserved record, ca. 4–4.55 billionyears ago). The Cambrian is the initial period of the Phanerozoic Eon; thus, allearlier time is commonly, if informally, referred to as “Precambrian.” See geo-logic timescale on page 2.

10 Chapter 1

Figure 1.3. Small shelly fossils in basal Cambrian rocks. (a) Anabarites trisulcatus,the tiny skeletons found in lowermost Cambrian beds along the Kotuikan River.These specimens come from rocks of comparable age in China. (b) Small shellyfossils of the types found higher in the Kotuikan cliffs; most of the forms seenhere are the skeletal spicules of chancellorids, enigmatic baglike animals foundwidely (and only) in Cambrian rocks. (Images courtesy of Stefan Bengtson)

(a)

(b)

feet above river level, rocks estimated to be about 525 million years oldcontain nearly one hundred different types of shells (figure 1.3b).Some, like the small cones in the cliff base, have a threefold symmetrythat differentiates them from most animals alive today. Others, how-ever, include small spiral shells that are recognizably the remains ofmollusks, bivalved skeletons formed by brachiopods, and, a littlehigher up, the segmented bodies of trilobites. Painstakingly collectedand described by Russian paleontologists, these fossils chronicle anapparently rapid unfolding of biological diversity in the Cambrianocean. In less than 20 million years, the seafloor was transfiguredfrom the alien to the (at least broadly) familiar. The same drama isrecorded in rocks of comparable age throughout the world, providingour earliest glimpses of the animals that have populated Earth’soceans ever since that time.

Charles Darwin couldn’t get this pattern out of his mind. One mightsuppose that Darwin, like his modern intellectual descendants, saw inthe fossil record a confirmation of his theory—the literal documentationof life’s evolution from the Cambrian to the present day. In fact, the twochapters devoted to geology in The Origin of Species are anything but cel-ebratory. On the contrary, they constitute a carefully worded apology inwhich Darwin argues that evolution by natural selection is correct de-spite an evident lack of support from fossils.

Darwin envisioned natural selection as a slow but continuous processby which biological lineages diverged and gradually grew more distinctfrom one another. Intermediate forms that link different species are rarein the modern world because selection inexorably acts against them. Butwhy don’t we see intermediates in time? Darwin’s expectation was thatsuccessive sedimentary beds should document the gradual transitionfrom one form, perhaps seen at the base of a cliff, to its morphologicallydistinct descendants at the top. That such series are rare he attributed tothe extreme imperfection of the fossil record.

The Origin is full of magisterial prose, words that are luminous aswell as illuminating. Darwin’s characterization of the geological recordis particularly striking: “a history of the world imperfectly kept, andwritten in a changing dialect; of this history we possess the last volumealone, relating only to two or three countries. Of this volume, only here

11In the Beginning?

and there a short chapter has been preserved; and of each page onlyhere and there a few lines.”

Darwin might well have embraced Julian Barnes’s description ofhuman remembrance as a metaphor for Earth’s geological memory:sedimentary rocks provide a succession of widely spaced snapshots, nota documentary film of our planetary history. At the local level observedin a roadside or cliff face, this view is well justified, and Darwin’s argu-ments seem strikingly modern. We understand today that sedimentaryrocks provide discontinuous records in which the boundary betweentwo layers may represent more time than the beds themselves. But thegeometry of sedimentary accumulation is more complex than the or-derly layer cake commonly seen in local outcrop. Viewed three-dimen-sionally, the layers pinch and swell like hills in a van Gogh landscape,thickening here and changing character, thinning there to feather edge.Time represented in one place by a hiatus between beds is recorded else-where by sediment accumulation. Viewed still more broadly, at anypoint in time such locally discontinuous records are forming in manybasins throughout the world. Thus, if we revisit Darwin’s metaphor forgeological history, we find that while his book is missing chapters, andthe chapters in hand are missing pages, we actually possess multiplecopies of the text and the parts that are missing vary from copy to copy.If we have a principle for interleaving the surviving pages, it is possibleto stitch together a composite record that, for the past 600 million years,at least, isn’t bad. The discipline of stratigraphy provides that principle,showing us that at least the broad biological patterns read from fossilsreflect evolution and not gross inadequacies of the rock record.

Biostratigraphers have known for more than a century that species com-monly appear in the fossil record fully formed, persist without muchchange for million of years, and then disappear. The sense conveyed bythis pattern, that form changes episodically rather than continuously,doesn’t arise because species appear only once, in a single bed. It is jus-tified because species commonly occur in many successive beds withlittle change, or at least little directional change, from bottom to top—apattern we can’t explain away as the product of sedimentary incom-pleteness (at least not without making assumptions that, in many cases,we know to be implausible). Recognizing this, Niles Eldredge andStephen Jay Gould argued in 1972 that it is this stratigraphic pattern of

12 Chapter 1

“punctuated equilibrium”—and not Darwin’s picture of gradualchange—that is most consistent with modern evolutionary theory. Mostnew species arise not from the insensibly gradual transformation oflarge populations but rather by the rapid differentiation of small, iso-lated populations at the periphery of the main group. The transforma-tions envisioned by Darwin occur, but they take place rapidly and lo-cally, after which populations of the descendant species are constrainedby natural selection to stay more or less the same until competitors orshifting environments spell their doom.

The everyday comings and goings of fossil species can be reconciledwith both evolutionary expectation and geological reality, but whatabout the spectacular pattern seen in the Kotuikan cliffs? How do we ex-plain this biological transformation of the oceans? If Darwin was con-cerned about the general lack of transitional forms in the fossil record, hewas truly disquieted by the apparently abrupt appearance of abundant,diverse, and anatomically complex animals in the oldest Cambrian beds:

There is another and allied difficulty which is much graver. I allude to the

manner in which numbers of species of the same group, suddenly appear

in the lowest known fossiliferous rocks. . . . The case must at present remain

inexplicable; and may be truly urged as a valid argument against the views

herein entertained.

Of course, the Origin does offer an explanation, and it is the one wemight expect—massive record failure at the base of the Cambrian Sys-tem. It isn’t, wrote Darwin, that no life preceded the fabulously compli-cated snails and trilobites of the Cambrian, but rather that their ances-tors’ record lay in older beds that are deeply buried, destroyed, orundiscovered. In another memorable passage, Darwin insisted that

if my theory be true, it is indisputable that before the lowest Silurian2 stra-

tum was deposited, long periods elapsed, as long as, or probably far longer

13In the Beginning?

2 In the mid–nineteenth century, debate about how to define and differentiatethe Cambrian and Silurian Systems remained unresolved. Darwin employedLondoner Roderick Murchison’s term Silurian for the oldest fossiliferous beds,despite the fact that his Cambridge mentor Adam Sedgwick had coined thename “Cambrian.” Not until 1879 did Charles Lapworth cut the Gordian knot,retaining Cambrian for the lower part of the disputed system, Silurian for itsupper portion, and Ordovician (after the Ordovici, an ancient and putatively ob-streperous Welsh tribe) for the contested interval of overlap.

than, the whole interval from the Silurian age to the present day; and that

during these vast, yet quite unknown periods of time, the world swarmed

with living creatures.

Back along the Kotuikan River, Misha and I sit on a gravel bar oppositethe cliffs and consider Darwin’s dilemma as we sip our evening tea.How could such complexity evolve so quickly? And if it didn’t reallyhappen so fast, where are the rocks that record life’s earlier history?

The sedimentary beds in the Kotuikan cliffs aren’t quite flat-lying; tec-tonic movements over millions of years have tilted them slightly down-ward to the west. Because of this, a hike toward the east, upstreamalong the river, reveals layers that sit ever farther below the level of theCambrian fossils. About fifteen miles up river—some 200 feet lower inthe sedimentary rock column—we encounter a sharp stratigraphicbreak, the base of the sedimentary package that includes the latest Pre-cambrian carbonate rocks and basal Cambrian animals (figure 1.4). Isthat the end of the sedimentary trail?

Not at all. What lies beneath these rocks is another, older succession of

14 Chapter 1

Figure 1.4. Upstream along the Kotuikan River, showing the angular uncon-formity between the latest Precambrian-Cambrian succession seen in figure 1.1and an older package of sedimentary rocks that lies beneath it.

sandstones, shales and carbonates. Set at an acute angle to the youngerbeds, this older package is itself more than 3,500 feet thick. The base ofthe Cambrian System is not the bottom of the stratigraphic record—notin northern Siberia, and not in many other places where tectonic cir-cumstance has preserved sedimentary rocks deposited one, two, oreven 3 billion years before Cambrian beds began to accumulate.

We can put Darwin’s conjecture to the test. Is the Cambrian Explosionthe beginning of biological history? Or is it the culmination of evolu-tionary events that extend much deeper into our planet’s past?

15In the Beginning?